ALIBI: Adversarial Legitimacy Injection in Binary Input against LLM Malware Analyzers
ALIBI embeds fake benign-narrative sections into binaries, flipping 30 of 35 malware samples to benign verdicts on Gemini 2.5 Pro.
ALIBI is a semantic cover-story attack against LLM-based malware analyzers that adds a small non-executed read-only section containing a false security-product narrative, without altering imports or executable behavior. On a frozen PE set, Gemini 2.5 Pro flipped 30 of 35 baseline-malicious samples to benign, while GPT-5.5 Pro and Claude Opus 4.7 showed substantial severity downgrades and confidence reductions. The attack transfers to ELF binaries (Gemini flips 16 of 40), and a verification-guided defense prompt only halves benign verdicts.
- Attacks LLM malware triage via false benign narrative, not instructions
- Gemini 2.5 Pro: 30/35 PE samples flipped to benign
- Transfers to ELF: 16/40 flipped on Gemini
- Defense prompt halves but does not eliminate misclassification
Full article185 words · extracted from arxiv.org · click to collapse
Large language models are being integrated into malware triage workflows as reasoning components that summarize static evidence and produce analyst-facing verdicts. This paper shows that the same reasoning capability introduces a new attack surface. We present ALIBI, a semantic cover story attack against frontier LLM-based malware analyzers. ALIBI adds a small, non-executed read-only section to a compiled binary, containing a coherent but false security product narrative, without altering imports or executable behavior. Instead of issuing direct instructions to the model, it reframes suspicious evidence as expected behavior of a benign endpoint security tool. On a frozen PE set of 50 malicious samples, the payload flips 30 of the 35 baseline-malicious samples to benign on Gemini 2.5 Pro, while GPT-5.5 Pro and Claude Opus 4.7 produce substantial severity downgrades with significant confidence reductions even when verdict labels are preserved. The attack transfers to ELF binaries, where Gemini flips 16 of 40. A verification-guided defense prompt roughly halves the benign verdicts, but 42.9 percent of malicious samples still reach benign. LLM malware analyzers therefore require provenance checks that separate verified facts from attacker-controlled claims, not narrative trust.
Text extracted automatically; images, tables and formatting may be missing. Original: https://arxiv.org/abs/2609.19722